on 11-17-2020 05:33 AM
Below you can find few best practices gathered from previous experiences in the field.
1. When correctness job runs, it generates "remediate duplicate task" if duplicates are found. The best practice is to remediate the issues in the task and not directly on the CI level (deleting the duplicate CI) for various reasons:4. Start implementing CMDB health as soon as you start the CMDB implementation. Configuring the different metrics and improving them little by little is more efficient than fixing hundreds of issues few monthes later.
5. Leverage the CMDB health groups to be able to load the different scores on a subset of CIs. In the main dashboard (CMDB Health - CMDB view) the scores can be loaded on all CIs or a child class. However the (CMDB Health - Group view) dashboard can be loaded on a CMDB group that represents a query on any attribute of the CI. This is an excellent way to slide and dice the scores and see them on different angles.
Hope this article accelerates your CMDB journey! Don't hesitate to share your thoughts and REX on this topic.
Regards,
e.
e.
Great article!!! I am just starting to dig into this. Have a few questions...
1. How can I run the "Completeness" for Only Applications (not every CI). I have been in CI Class Manager for App Health, but how can I ensure that's the only CI Class.
2. Is there any training video on this CMDB Dashboard?
I created a Health Inclusion Rule, but the main CMDB Health Scorecard remains the same, it did not filter out other CIs. According to everything I've red, if you configure Health Inclusion Rules, they control what is shown in the CMDB Health View Scorecard. Am I missing something?
Some time has passed, and i don't know if the questions are still valid.
@Marc Schlank1, try creating CMDB Groups (type Health) for the sets of CIs you want to monitor (eg. Applications). Then display CMDB Health - Group View and pick the group you want. This has worked for us.
@Carlis Ragland, we created health inclusion rules at the top level: Configuration Item (cmdb_ci). I've tried both blacklisting and whitelisting, and they're about equally complicated. They do seem to work, though. After I exclude the classes we don't want, the total population under review is smaller and looks accurate.
The path into health inclusion rules is a little strange. Initially, there are none. Then after specifying some conditions (typically "Class is X" and/or "Class is not Y"), SN translates those into separate rules for each of the main 6 metrics. For further modification, you have to modify each of those, or wipe them all out and start again from nothing. A little inconvenient.
What is your experience? Are you getting CMDB Health to behave as you expect in response to the inclusion rules?
Hello ,
In the screenshot ,I can see completeness is 41/3231 ,What is this 3231 ?
I tried using inclusion rule on windows class it gave me number of 23/24 though I have 3k CI in windows class
@rajat11 i believe that is Scope of CI's that have been used to generate the completeness.